1. Field of the Invention
The present invention relates to a communication field, and more particularly to an interference cancellation method and apparatus.
2. Description of the Prior Art
In a DS-CDMA (Direct Sequence-Code Division Multiple Access) system, CDMA (Code Division Multiple Access) technologies may be mainly adopted. Due to the propagation time delays of different signals being different and the presence of scrambling codes, spreading code sets adopted by respective signals are not completely orthogonal. Such interference caused by a non-zero cross-correlation coefficient is usually called MAI (Multiple Access Interference).
In a CDMA system, a MF (Matched Filter) (matched filter theory being adopted by a traditional Rake receiver) or a MUD (Multi User Detector) is usually adopted to restore the data before spreading and scrambling. However, the traditional matched filter apparatus may not efficiently suppress the MAI, while the MUD apparatus may preferably eliminate the impact caused by the MAI.
In a TD-SCDMA (Time Division-Synchronous Code Division Multiple Access) system, the MUD apparatus may adopt a JD (joint Detector) apparatus, which is a kind of linear MUD apparatus. Due to the operation of system matrix inversion being required to be implemented, the dimension of the system matrix will increase as a SF (Spread Factor) adopted by the CDMA system is large, the length of the scrambling code is long or the number of the interference user is large, which results in that the complexity of the operation for the matrix inversion increases and the performance of the apparatus is reduced, and the efficiency of the process for cancelling the interference is low. Meanwhile, the existence of a large number of strong interference signals in the system may impact the output performance of the whole equalizer.
No effective resolution has yet been proposed for the problem of low efficiency of the process for cancelling the interference if the adopted SF is large, the length of the scrambling code is long or the number of the interference user is large in the related art. In addition, no effective resolution has yet been proposed for the problem of low system demodulation performance caused by the large number of strong interference signals.